A Generalized Projective Quasi-Multiplier-Strongly Subfeasible- Directions Method for Nonlinear Optimization Problems

摘要

In this paper, algorithms for solving optimization problems with general nonlinearly equality and inequality constraints are discussed, the new concepts of quasi-Kuhn Tucker point and quasi-multiplier methods are put forward. With the help of generalized projection and strongly sub-feasible directions methods, we present a new algorithm which is called quasi-multiplier-strongly subfeasible directions method. Under mild assumptions, the algorithm is proved to converge to the quasi-Kuhn Tucker point of the solved problem.